DESCRIPTION: (Applicant's Abstract) Malignant brain tumors continue to increase in incidence in the US, and currently are the most common solid tumors of childhood and adolescence. Unfortunately, however, brain tumors remain among the most therapeutically intractable of human tumors, and long term survivors are rare among patients with highly anaplastic astrocytomas or glioblastoma multiforme. A major cause of failure of brain tumor therapy, as in most other human cancers, is drug resistance, and much effort has been devoted towards understanding the cellular and molecular mechanisms that underly it. These studies have shown that drug resistance mechanisms often involve the dysregulation of genes, many of which are involved in normal cellular processes, such as metabolism, transport, DNA repair and cell cycle progression. One of the best characterized of these mechanisms is that of GST-pi over-expression. This application is founded on two significant recent findings from the applicant's laboratory related to the GST-pi gene. The first is that in gliomas, GST-pi over-expression is associated with drug resistance, malignant progression and poor patient survival. Secondly, he has made the potentially very important discovery that the human GST-pi gene locus is polymorphic and contains, at least, three allelic GST-pi gene variants. One of these variants, hGSTP1*C, is more frequently present in gliomas than in normal cells/tissues. The applicant has cloned the variant cDNAs and shown the encoded proteins to be structurally and functionally different. These findings are having a significant impact in the field of GST research. The primary goal of this application is to examine the influence of this newly discovered GST-pi genetic polymorphism on drug resistance in human gliomas and to determine whether specific GST-pi genotype/phenotypes are associated with differential therapeutic outcome and in patient survival. The Specific aims are: 1) To determine by molecular dynamic modeling, the differential binding affinities of anticancer agents to the active sites of proteins encoded by GST-pi allelic gene variants and correlate these with the differential abilities of the GST-pi proteins to inactivate anticancer agents; 2) To determine whether different GST-pi gene variants confer different levels of drug resistance to malignant glioma cells; 3) To determine whether GST-pi allelotype is related to the level of in vitro drug resistance of gliomas, and with in vivo response to therapy and survival of glioma patients following chemotherapy; and, 4) To determine whether down-regulation of GST-pi gene expression in gliomas that express different GST-pi gene variants will differentially affect drug resistance. The applicant believes that this application is well-focused and has a significant degree of novelty, with respect to the hypothesis, preliminary data and experimental techniques to be used. He believes the results are likely to make important and critical contributions to understanding the cellular, molecular and genetic mechanisms involved in GST-pi mediated drug resistance in human gliomas that will be applicable to many other human tumor types for which GST-pi over-expression has been shown to be an important determinant of drug resistance and failure of patients to respond to therapy.